panic in propagate_priority w/ postgresql under heavy load

[Koen Martens]

Vinod Kashyap wrote:
> You seem to be booting off of a 9000 (twa) controller and not 7000/8000
> (twe).
> It could be because of a 9000 firmware bug that you are not being able
> to
> get the dump.  The firmware wrongly interprets physical address 0x0 as
> invalid
> during dumps, and fails the operations.  This bug will be fixed in
> future
> firmware releases.

Indeed am I booting of twa, swap is also on there. Just got back
from vacation, we did have another panic. The box was booted into
single user mode right after that, after which an image of the swap
partition was made with dd. When I got back, i turned of swap
momentarily and dd'ed the image back on the swap partition, after
which i ran savecore. However, savecore reports 'no dumps found'.
With the -f option it also says: 'unable to force dump - bad magic'

(Note: swap is 2048mb, physical memory is also 2048mb).

So basically, what i'm left with now is a production server that
crashes every x days, possibly resulting in some database
corruption, and no way to obtain more info about the crash than i
have already provided...

I will try and install a comparable setup on a spare box, without
the twa (plain IDE) and see if i can reproduce something, if i can
find some time to do this in.

I will also try bumping the witness limits. What is this witness
business anyway, what does it output and/or where do i RTFM about it?

Gr,

Koen

-- 
K.F.J. Martens, Sonologic, http://www.sonologic.nl/
Networking, hosting, embedded systems, unix, artificial intelligence.
Public PGP key: http://www.metro.cx/pubkey-gmc.asc
Wondering about the funny attachment your mail program
can't read? Visit http://www.openpgp.org/